Search: | Resources

newport business swansea business south wales business directory cardiff business directory cardiff business advice wales business consulting

PyPy - Automatic Generation of VMs for Dynamic Languages

internetmarketingsoftwaretools.com videos
internetmarketingsoftwaretools.com videos internetmarketingsoftwaretools.com videos
internetmarketingsoftwaretools.com videos

Google Tech TalksNovember, 14 2007In Search of Speed and Flexibility. PyPy and the Art of Generating Virtual MachinesWe all want our favorite dynamic language to be faster, to be even moredynamic and to pick up nifty ideas from its peers.But this is so much work!There ought to be a better way to implement dynamic languages thatenables growth and necessary change with less effort and more ease.Writing yet another interpreter in C or even Java is not it.PyPy is a framework to implement dynamic languages that embraces thischallenge. It allows us to write such languages in the form of simpleinterpreters, expressed in a rich subset of Python. This means thatthe resulting implementations can can be evolved with reasonableeffort.PyPy then generates a variety of virtual machines (VMs) from a singlesource. We can therefore target a wide range of environments,including C/Posix, Java, or .NET. This avoids the major source offragmentation within a dynamic language community -- the need for aseparate language implementation for each virtual machine. It alsomeans that languages as distinct as Python, Prolog and Smalltalk canshare the bulk of our compiler machinery.The flexible toolchain which is necessary for virtual machinegeneration turns out to be good for other things as well. Traditionalimplementations for dynamic languages have to make some very harddecisions early, which then become entrenched in the whole codebase,making it difficult or impossible to change later. It can be nice toexperiment with different garbage collectors, but not if you have towrite a complete new implementation of your favourite language foreach one. With PyPy you don't have to. You can also generate andtailor a JIT-compiler for the language at hand, putting to rest thenotion that flexibility must come at the expense of speed.Speaker: Samuele PedroniSpeaker: Armin RigoSpeaker: Jacob HallénSpeaker: Laura Creighton

Channel: People & Blogs
Uploaded: November 30, 1999 at 12:00 am
Author: googletechtalks

Length: 58:12
Rating: 3.91
Views: 8209

Tags: education  engedu  google  googletechtalks  talk  talks  techtalk  techtalks  

Video Url:


Embed Code:

Video Comments

cybernytrix (November 30, 1999 at 12:00 am)
The presentation is totally unimpressive. I also do not believe that they have anyhting that is useful in the real world. There is not way in hell you can write an interpreter in RPython and make it run as fast as CPython or Jython *ever*. Remember they use JIT, so you must use Psyco to make a fair comparison.Smoke on...
klined (November 30, 1999 at 12:00 am)
pypy is the most ambitious project any language has ever had
collodiallofi (November 30, 1999 at 12:00 am)
I do not consider any of the speakers to be remotely hot
pbrinkhall (November 30, 1999 at 12:00 am)
The audio sux!
foundtheory (November 30, 1999 at 12:00 am)
The majority of the talk (and most interesting parts!) is free of audio problems...Great talk, a lot of good information on PyPy
robinsto (November 30, 1999 at 12:00 am)
After 7 min 30 s, the audio is good.
polukaks (November 30, 1999 at 12:00 am)
actually, the audio is just fine after some initial troubles.
alunufal (November 30, 1999 at 12:00 am)
Very bad audio. Too low and most of the time ininteligible.
davdunc (November 30, 1999 at 12:00 am)
Audio is, "really, really, really" bad.
demian0311 (November 30, 1999 at 12:00 am)
audio is bad at times

SouthWalesBusinessWebsites © 2007 All Rights Reserved.